Reproduce “ElasticFormer: Detecting Objects in HRW Shots via Elastic Computing Vision Transformer” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Wenxi Li, Jingchen Huang, Chenyang Lyu, Moran Liu, Haozhe Lin, Guiguang Ding, Yuchen Guo
ElasticFormer adapts high-resolution object detection by estimating each image’s foreground coverage and selecting a corresponding number of high-scoring windows for local processing. On PANDA’s standardized 1280 × 800 test crops, DINO+ElasticFormer achieves AP50 of 0.806 with 13.13 GFLOPs, versus 0.780 and 75.71 GFLOPs for DINO+SparseFormer. This shows reduced operations and higher accuracy, not measured inference speed.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.